A Long Skip Connection for Enhanced Color Selectivity in CNN Architectures

Some recent studies show that filters in convolutional neural networks (CNNs) have low color selectivity in datasets of natural scenes such as Imagenet. CNNs, bio-inspired by the visual cortex, are characterized by their hierarchical learning structure which appears to gradually transform the repres...

Volledige beschrijving

Bibliografische gegevens
Hoofdauteurs: Oscar Sanchez-Cesteros, Mariano Rincon, Margarita Bachiller, Sonia Valladares-Rodriguez
Formaat: Artikel
Taal:English
Gepubliceerd in: MDPI AG 2023-08-01
Reeks:Sensors
Onderwerpen:
Online toegang:https://www.mdpi.com/1424-8220/23/17/7582